One of the most important problems in carcinogenesis is the analysis of the relationship between exposure to radiation and chemical carcinogens and subsequent altered gene expression. We have developed several powerful strategies to analyze this problem at the molecular level. Recently we have shown that ultraviolet radiation (UV) can convert S49 mouse cells from Cd sensitive (Cds) to Cd resistant (Cdr) and that in many of the Cdr lines the metallothione in-I (MT-I) gene has been activated. Experiments are planned to determine if direct acting chemical carcinogens can also activate the quiescent MT-I gene in S49 cells. The role of repair related processes (e.g., undermethylation in repair patches) will be assessed by studying this phenomenon in MT negative, Cds, repair-competent and repair-deficient CHO cells. During MT-I activation one of the two copies in S49 cells becomes unmethylated. One may take advantage of this finding to clone the activated and quiescent alleles and compare them by sequencing and in functional assays (transient expression, transformation) with or without in vitro methylation. These and related studies should help determine the mechanism of MT-I activation; i.e., the relation between mutational events and non-mutational events (changes in methylation patterns). The structure of a related gene, metallothionein-II (MT-II), is now available, and coordinate activation of these functionally related and physically linked genes may be studied. To assess the generality of gene activation by carcinogens we are developing a system to evaluate reactivation of thymidine kinase (TK) in cells in which the gene has been inactivated. These studies will involve assessment of changes in gene structure and TK mRNA synthesis in lines converted to TK+ by carcinogens. In addition, we have developed a strategy to clone the activated alleles. To evaluate the role of carcinogen-induced activation of genes in neoplastic transformation, we plan to analyze changes in expression and genomic arrangement of known oncogenes in cultured cells transformed to anchorage independence by treatment with carcinogens. Because the detailed characterization of oncogenes is rapidly becoming available, mechanistic studies similar to those described for MT-I, MT-II, and TK should be possible.